Skip to content

Commit

Permalink
Remove dev and main branch from workflows. (ROCm#404)
Browse files Browse the repository at this point in the history
* Remove `dev` and `main` branch from workflows.

Update links in documentation.

Signed-off-by: David Galiffi <David.Galiffi@amd.com>

* `amd-staging` -> `amd-mainline` in docs

Signed-off-by: Peter Jun Park <peter.park@amd.com>

---------

Signed-off-by: David Galiffi <David.Galiffi@amd.com>
Signed-off-by: Peter Jun Park <peter.park@amd.com>
Co-authored-by: Peter Jun Park <peter.park@amd.com>
Signed-off-by: xuchen-amd <xuchen@amd.com>
  • Loading branch information
2 people authored and xuchen-amd committed Oct 17, 2024
1 parent 8cf32d3 commit 91654cf
Showing 14 changed files with 24 additions and 24 deletions.
2 changes: 1 addition & 1 deletion .github/workflows/docs.yml
Original file line number Diff line number Diff line change
@@ -2,7 +2,7 @@ name: Documentation

on:
push:
branches: [ main, amd-mainline ]
branches: [ amd-mainline ]
paths:
- 'docs/archive/docs-2.x/**'
- 'docs/archive/docs-1.x/**'
4 changes: 2 additions & 2 deletions .github/workflows/formatting.yml
Original file line number Diff line number Diff line change
@@ -3,9 +3,9 @@ name: Formatting

on:
push:
branches: [ main, dev, amd-mainline, amd-staging ]
branches: [ amd-mainline, amd-staging ]
pull_request:
branches: [ main, dev, amd-mainline, amd-staging ]
branches: [ amd-mainline, amd-staging ]

concurrency:
group: ${{ github.workflow }}-${{ github.ref }}
2 changes: 1 addition & 1 deletion .github/workflows/mi-rhel9.yml
Original file line number Diff line number Diff line change
@@ -2,7 +2,7 @@ name: mi-rhel9

on:
push:
branches: [ main, amd-mainline ]
branches: [ amd-mainline ]

# Allows manual execution
workflow_dispatch:
4 changes: 2 additions & 2 deletions .github/workflows/rhel-8.yml
Original file line number Diff line number Diff line change
@@ -5,9 +5,9 @@ name: RHEL 8
# Controls when the workflow will run
on:
push:
branches: [ main, dev, amd-mainline, amd-staging ]
branches: [ amd-mainline, amd-staging ]
pull_request:
branches: [ main, dev, amd-mainline, amd-staging ]
branches: [ amd-mainline, amd-staging ]

# Allows you to run this workflow manually from the Actions tab
workflow_dispatch:
2 changes: 1 addition & 1 deletion .github/workflows/tarball.yml
Original file line number Diff line number Diff line change
@@ -2,7 +2,7 @@ name: tarball

on:
push:
branches: [ main, amd-mainline ]
branches: [ amd-mainline ]
pull_request:

concurrency:
4 changes: 2 additions & 2 deletions .github/workflows/ubuntu-jammy.yml
Original file line number Diff line number Diff line change
@@ -4,9 +4,9 @@ name: Ubuntu 22.04

on:
push:
branches: [ main, dev, amd-mainline, amd-staging ]
branches: [ amd-mainline, amd-staging ]
pull_request:
branches: [ main, dev, amd-mainline, amd-staging ]
branches: [ amd-mainline, amd-staging ]

# Allows you to run this workflow manually from the Actions tab
workflow_dispatch:
2 changes: 1 addition & 1 deletion docs/archive/docs-1.x/analysis.md
Original file line number Diff line number Diff line change
@@ -171,7 +171,7 @@ $ omniperf analyze -p workloads/vcopy/mi200/ --list-metrics gfx90a
├─────────┼─────────────────────────────┤
...
```
2. Choose your own customized subset of metrics with `-b` (a.k.a. `--metric`), or build your own config following [config_template](https://github.com/ROCm/omniperf/blob/main/src/omniperf_analyze/configs/panel_config_template.yaml). Below shows how to generate a report containing only metric 2 (a.k.a. System Speed-of-Light).
2. Choose your own customized subset of metrics with `-b` (a.k.a. `--metric`), or build your own config following [config_template](https://github.com/ROCm/omniperf/blob/amd-mainline/src/omniperf_analyze/configs/panel_config_template.yaml). Below shows how to generate a report containing only metric 2 (a.k.a. System Speed-of-Light).
```shell-session
$ omniperf analyze -p workloads/vcopy/mi200/ -b 2
--------
2 changes: 1 addition & 1 deletion docs/archive/docs-2.x/analysis.md
Original file line number Diff line number Diff line change
@@ -181,7 +181,7 @@ Analysis mode = cli
2.1.30 -> L1I Fetch Latency
...
```
3. Choose your own customized subset of metrics with `-b` (a.k.a. `--block`), or build your own config following [config_template](https://github.com/ROCm/omniperf/blob/main/src/omniperf_analyze/configs/panel_config_template.yaml). Below shows how to generate a report containing only metric 2 (a.k.a. System Speed-of-Light).
3. Choose your own customized subset of metrics with `-b` (a.k.a. `--block`), or build your own config following [config_template](https://github.com/ROCm/omniperf/blob/amd-mainline/src/omniperf_analyze/configs/panel_config_template.yaml). Below shows how to generate a report containing only metric 2 (a.k.a. System Speed-of-Light).
```shell-session
$ omniperf analyze -p workloads/vcopy/MI200/ -b 2
--------
16 changes: 8 additions & 8 deletions docs/archive/docs-2.x/performance_model.md
Original file line number Diff line number Diff line change
@@ -2178,7 +2178,7 @@ A good discussion of coarse and fine grained memory allocations and what type of
(VALU_inst_mix_example)=
## VALU Arithmetic Instruction Mix

For this example, we consider the [instruction mix sample](https://github.com/ROCm/omniperf/blob/dev/sample/instmix.hip) distributed as a part of Omniperf.
For this example, we consider the [instruction mix sample](https://github.com/ROCm/omniperf/blob/amd-mainline/sample/instmix.hip) distributed as a part of Omniperf.

```{note}
This example is expected to work on all CDNA accelerators, however the results in this section were collected on an [MI2XX](2xxnote) accelerator
@@ -2269,7 +2269,7 @@ shows that we have exactly one of each type of VALU arithmetic instruction, by c
(Fabric_transactions_example)=
## Infinity-Fabric(tm) transactions

For this example, we consider the [Infinity Fabric(tm) sample](https://github.com/ROCm/omniperf/blob/dev/sample/fabric.hip) distributed as a part of Omniperf.
For this example, we consider the [Infinity Fabric(tm) sample](https://github.com/ROCm/omniperf/blob/amd-mainline/sample/fabric.hip) distributed as a part of Omniperf.
This code launches a simple read-only kernel, e.g.:

```c++
@@ -2826,7 +2826,7 @@ On an AMD [MI2XX](2xxnote) accelerator, for FP32 values this will generate a `gl
(flatmembench)=
### Global / Generic (FLAT)

For this example, we consider the [vector-memory sample](https://github.com/ROCm/omniperf/blob/dev/sample/vmem.hip) distributed as a part of Omniperf.
For this example, we consider the [vector-memory sample](https://github.com/ROCm/omniperf/blob/amd-mainline/sample/vmem.hip) distributed as a part of Omniperf.
This code launches many different versions of a simple read/write/atomic-only kernels targeting various address spaces, e.g. below is our simple `global_write` kernel:

```c++
@@ -2976,7 +2976,7 @@ The assembly in these experiments were generated for an [MI2XX](2xxnote) acceler
Next, we examine a generic write.
As discussed [previously](Flat_design), our `generic_write` kernel uses an address space cast to _force_ the compiler to choose our desired address space, regardless of other optimizations that may be possible.

We also note that the `filter` parameter passed in as a kernel argument (see [example](https://github.com/ROCm/omniperf/blob/dev/sample/vmem.hip), or [design note](Flat_design)) is set to zero on the host, such that we always write to the 'local' (LDS) memory allocation `lds`.
We also note that the `filter` parameter passed in as a kernel argument (see [example](https://github.com/ROCm/omniperf/blob/amd-mainline/sample/vmem.hip), or [design note](Flat_design)) is set to zero on the host, such that we always write to the 'local' (LDS) memory allocation `lds`.

Examining this kernel in the VMEM Instruction Mix table yields:

@@ -3339,7 +3339,7 @@ Next we examine the use of 'Spill/Scratch' memory.
On current CDNA accelerators such as the [MI2XX](2xxnote), this is implemented using the [private](mspace) memory space, which maps to ['scratch' memory](https://llvm.org/docs/AMDGPUUsage.html#amdgpu-address-spaces) in AMDGPU hardware terminology.
This type of memory can be accessed via different instructions depending on the specific architecture targeted. However, current CDNA accelerators such as the [MI2XX](2xxnote) use so called `buffer` instructions to access private memory in a simple (and typically) coalesced manner. See [Sec. 9.1, 'Vector Memory Buffer Instructions' of the CDNA2 ISA guide](https://www.amd.com/system/files/TechDocs/instinct-mi200-cdna2-instruction-set-architecture.pdf) for further reading on this instruction type.

We develop a [simple kernel](https://github.com/ROCm/omniperf/blob/dev/sample/stack.hip) that uses stack memory:
We develop a [simple kernel](https://github.com/ROCm/omniperf/blob/amd-mainline/sample/stack.hip) that uses stack memory:
```c++
#include <hip/hip_runtime.h>
__global__ void knl(int* out, int filter) {
@@ -3404,7 +3404,7 @@ Here we see a single write to the stack (10.3.6), which corresponds to an L1-L2
(IPC_example)=
## Instructions-per-cycle and Utilizations example

For this section, we use the instructions-per-cycle (IPC) [example](https://github.com/ROCm/omniperf/blob/dev/sample/ipc.hip) included with Omniperf.
For this section, we use the instructions-per-cycle (IPC) [example](https://github.com/ROCm/omniperf/blob/amd-mainline/sample/ipc.hip) included with Omniperf.

This example is compiled using `c++17` support:

@@ -3824,7 +3824,7 @@ Finally, we note that our branch utilization (11.2.5) has increased slightly fro

## LDS Examples

For this example, we consider the [LDS sample](https://github.com/ROCm/omniperf/blob/dev/sample/lds.hip) distributed as a part of Omniperf.
For this example, we consider the [LDS sample](https://github.com/ROCm/omniperf/blob/amd-mainline/sample/lds.hip) distributed as a part of Omniperf.
This code contains two kernels to explore how both [LDS](lds) bandwidth and bank conflicts are calculated in Omniperf.

This example was compiled and run on an MI250 accelerator using ROCm v5.6.0, and Omniperf v2.0.0.
@@ -4037,7 +4037,7 @@ The bank conflict rate linearly increases with the number of work-items within a
## Occupancy Limiters Example


In this [example](https://github.com/ROCm/omniperf/blob/dev/sample/occupancy.hip), we will investigate the use of the resource allocation panel in the [Workgroup Manager](SPI)'s metrics section to determine occupancy limiters.
In this [example](https://github.com/ROCm/omniperf/blob/amd-mainline/sample/occupancy.hip), we will investigate the use of the resource allocation panel in the [Workgroup Manager](SPI)'s metrics section to determine occupancy limiters.
This code contains several kernels to explore how both various kernel resources impact achieved occupancy, and how this is reported in Omniperf.

This example was compiled and run on a MI250 accelerator using ROCm v5.6.0, and Omniperf v2.0.0:
2 changes: 1 addition & 1 deletion docs/conf.py
Original file line number Diff line number Diff line change
@@ -55,7 +55,7 @@

# frequently used external resources
extlinks = {
"dev-sample": ("https://github.com/ROCm/omniperf/blob/dev/sample/%s", "%s"),
"dev-sample": ("https://github.com/ROCm/omniperf/blob/amd-mainline/sample/%s", "%s"),
"prod-page": (
"https://www.amd.com/en/products/accelerators/instinct/%s.html",
"%s",
2 changes: 1 addition & 1 deletion docs/how-to/analyze/cli.rst
Original file line number Diff line number Diff line change
@@ -186,7 +186,7 @@ Walkthrough
3. Choose your own customized subset of metrics with the ``-b`` (or ``--block``)
option. Or, build your own configuration following
`config_template <https://github.com/ROCm/omniperf/blob/main/src/omniperf_soc/analysis_configs/panel_config_template.yaml>`_.
`config_template <https://github.com/ROCm/omniperf/blob/amd-mainline/src/omniperf_soc/analysis_configs/panel_config_template.yaml>`_.
The following snippet shows how to generate a report containing only metric 2
(:doc:`System Speed-of-Light </conceptual/system-speed-of-light>`).

2 changes: 1 addition & 1 deletion docs/how-to/profile/mode.rst
Original file line number Diff line number Diff line change
@@ -38,7 +38,7 @@ Run ``omniperf profile -h`` for more details. See
Profiling example
-----------------

The `<https://github.com/ROCm/omniperf/blob/main/sample/vcopy.cpp>`__ repository
The `<https://github.com/ROCm/omniperf/blob/amd-mainline/sample/vcopy.cpp>`__ repository
includes source code for a sample GPU compute workload, ``vcopy.cpp``. A copy of
this file is available in the ``share/sample`` subdirectory after a normal
Omniperf installation, or via the ``$OMNIPERF_SHARE/sample`` directory when
Original file line number Diff line number Diff line change
@@ -623,7 +623,7 @@ manner. See
for further reading on this instruction type.

We develop a `simple
kernel <https://github.com/ROCm/omniperf/blob/dev/sample/stack.hip>`__
kernel <https://github.com/ROCm/omniperf/blob/amd-mainline/sample/stack.hip>`__
that uses stack memory:

.. code-block:: cpp
2 changes: 1 addition & 1 deletion docs/tutorial/profiling-by-example.rst
Original file line number Diff line number Diff line change
@@ -7,7 +7,7 @@ Profiling by example
********************

The following examples refer to sample :doc:`HIP <hip:index>` code located in
:fab:`github` :dev-sample:`ROCm/omniperf/blob/dev/sample <>` and distributed
:fab:`github` :dev-sample:`ROCm/omniperf/blob/amd-mainline/sample <>` and distributed
as part of Omniperf.

.. include:: ./includes/valu-arithmetic-instruction-mix.rst

0 comments on commit 91654cf

Please sign in to comment.